HORTSCIENCE 48(12):1452–1456. 2013.
Chlorophyll, Carotenoid, and Visual Color Rating of Japanese-cedar Grown in the Southeastern United States Ryan N. Contreras1,4 Department of Horticulture, Oregon State University, 4017 Agricultural and Life Sciences Building, Corvallis, OR 97331 John M. Ruter2 Allan Armitage Professor of Horticulture, Department of Horticulture, University of Georgia, 221 Hoke Smith, Athens, GA 30602 James S. Owen, Jr.1 Department of Horticulture, Virginia Polytechnic Institute and State University, Hampton Roads Agricultural Research and Extension Center, 1444 Diamond Springs Road, Virginia Beach, VA 23455 Andy Hoegh3 Laboratory for Interdisciplinary Statistical Analysis, Department of Statistics, Virginia Polytechnic Institute and State University, 406-A Hutcheson Hall, Blacksburg, VA 24061 Additional index words. Cryptomeria japonica, photoinhibition, pigments, winter-browning Abstract. Japanese-cedar has been underused in landscapes of the United States until recent years. There are now over 100 cultivars, many of which are grown in the southeast of the United States. Performance of cultivars has been described from U.S. Department of Agriculture (USDA) Zone 6b to USDA Zone 7b; however, there are no reports on how cultivars perform in USDA Zone 8. The current study was conducted to measure chlorophyll a, chlorophyll b, total chlorophyll, and carotenoid content and assign visual color ratings to determine if there was a relationship between pigment values and perceived greenness, which generally is regarded as a desirable and potentially heritable trait. Total chlorophyll (P = 0.0051), carotenoids (P = 0.0266), and the ratio of total chlorophyll to carotenoids (P = 0.0188) exhibited a positive relationship with greenness after accounting for season and tree effects. In contrast, the ratio of chlorophyll a to chlorophyll b did not have an effect on greenness. There was a linear relationship between total chlorophyll and carotenoid regardless of season (summer R2 = 0.94; winter R2 = 0.88) when pooled across 2 years. The observed correlation between chlorophyll and carotenoid content suggests they could be used interchangeably as predictors of greenness. There were large differences in rainfall between the 2 years that may have resulted in additional variation. Furthermore, the climate in which the evaluation was conducted differs greatly from the native distribution of japanese-cedar occurring in China and Japan. Japanese-cedar [Cryptomeria japonica (L.f.) D. Don] is a variable conifer that grows up to 60 m tall in its native range. Wild-type specimens are conical when young and become cylindrical with age (Eckenwalder, 2009). Japanese-cedar traditionally has been used as screening or specimen plantings; however, there are a large number of cultivars displaying varying forms and growth rates (Rouse et al., 2000; Tripp, 1993) that may be used in
Received for publication 16 Sept. 2013. Accepted for publication 28 Oct. 2013. We thank Nancy Hand, Bruce Tucker, and Zhibing Xu for their technical assistance. 1 Assistant Professor. 2 Professor. 3 PhD Candidate. 4 To whom reprint requests should be addressed; e-mail
[email protected].
1452
diverse landscape situations. There are estimates of over 100 different ornamental cultivars (R. Determann, personal communication; Erhardt, 2005) with 45 of these grown in the eastern United States (Rouse et al., 2000). Japanese-cedars are native to the warmtemperate zones of south China and Japan. In Japan, they generally are limited to northfacing slopes that receive 180 to 300 cm of rainfall per year (Tsukada, 1967). Japanesecedars perform well under a number of environmental and soil conditions including the hot, humid summers and heavy clay soils of the southeast United States (Tripp and Raulston, 1992). As a result of this fact, japanese-cedars have been promoted as an alternative to Leyland cypress [·Cuprocyparis leylandii (A.B.Jacks. & Dallim.) Farjon (Farjon et al., 2002)], on which numerous problems now occur including bagworms (Thyridopteryx
ephemeraeformis Haworth) (Lemke et al., 2005), fungal cankers caused by Seiridium Nees ex Link spp. and Botryosphaeria dothidea (Moug.) Ces. et De Not., and cercospora needle blight (Cercosporidium sequoiae Ellis and Everh.) (Martinez et al., 2009). Japanesecedars exhibit less susceptibility to bagworm infestations, and cultivars are available that have reduced interior branch death (Tripp and Raulston, 1992); however, the species is not problem-free. Redfire (Phyllosticta aurea C.Z. Wang) is a fungal pathogen that can attack stressed japanese-cedars and cause stem death, particularly on older foliage (Cox and Ruter, 2013; Tripp, 2005). Also, as Dirr (2009) notes, there is not a fast-growing, treelike cultivar that remains green during winter. Winter browning in japanese-cedar is often unsightly and undesirable to consumers, which may have contributed to why it has remained underused in landscapes. Winter browning in japanese-cedar occurs through the conversion of chloroplasts to chromoplasts during winter (Ida, 1981). This transition takes place only in sun-exposed leaves during periods of low temperature, indicating that photoinhibition likely plays a role (Han and Mukai, 1999; Ida, 1981). Plants have several mechanisms to cope with excess light during periods of low temperature when Calvin cycle activity is limiting, including reduction of chlorophyll, pH-dependent xanthophyll cycle, increased levels of carotenoids, and production of antioxidants or reactive oxygen species (ROS) scavenging enzymes. During winter, japanese-cedar has been shown to demonstrate two principle mechanisms to deal with excess light energy. The amount of chlorophyll decreases during winter (Han and Mukai, 1999; Ida, 1981) in both sunand shade-exposed leaves (Han et al., 2004). This occurs in both wild-type and nonbrowning mutants (Han et al., 2003), thus reducing the amount of energy absorbed. The other mechanism is the conversion of chloroplasts in sun-exposed leaves to rhodoxanthin-containing chromoplasts during winter to dissipate excess light energy as heat (Ida, 1981). In a study on 15-year-old japanese-cedar trees in Shizuoka Prefecture, Japan, Han et al. (2004) reported accumulation of rhodoxanthin in sun-exposed leaves beginning in January, reaching maximum levels in February, decreasing significantly in March, and falling to zero by April. However, timing of discoloring in winter and restoration in spring is highly variable and location-specific (personal observation). Han et al. (2003) demonstrated that wild-type leaves that accumulated rhodoxanthin maintained higher levels of photosynthesis with lower levels of zeaxanthin- and antheraxanthin-dependent thermal dissipation than mutants that remained green all winter. The proposed role of rhodoxanthin is to intercept a portion of incident light to help maintain an appropriate balance among light absorption, thermal dissipation, and photosynthesis (Han et al., 2003). Japanese-cedar also accumulated substantial levels of xanthophyll cycle pigments and lutein during HORTSCIENCE VOL. 48(12) DECEMBER 2013
winter (Han et al., 2003; Han and Mukai, 1999). The overarching objective of the current study was to identify an early predictor of winter foliage color (resistance to leaf browning) in japanese-cedar as a screening tool for identifying superior selections. Specifically, we assessed if quantitating pigments such as total chlorophyll (Ca+b), ratio of chlorophyll a (Ca):chlorophyll b (Cb), total carotenoids (Cx+c), and ratio of (Ca+b):(Cx+c) exhibited a strong relationship with greenness as measured by visual color rating at the University of Georgia Tifton Campus (USDA Zone 8b; USDA-ARS, 2012). Materials and Methods Plant material and growing conditions. Single plants of the following 12 taxa of japanese-cedar were randomly planted in 1997: ‘Araucariodes’, ‘Ben Franklin’, ‘Black Dragon’, ‘Cristata’, ‘Gyokruga’, ‘Rasen’, ‘Sekkan’, ‘Tansu’, ‘Tarheel Blue’, var. sinensis, ‘Yaku’, and ‘Yoshino’. These taxa were not meant to provide an evaluation of each per se as a result of single replicates; rather, they were used to provide a general sampling of the genotypic and phenotypic diversity observed in landscape forms of japanesecedar. Plants were maintained in field plots at the University of Georgia Tifton Campus (lat. 3149’ N, long. 8353’ W; USDA Zone 8b). Field soil was a Tifton loamy sand (fineloamy, siliceous, thermic Plinthic Paleudult), pH 5.2. Plots were fertilized in March every year after planting at a rate of 56 kg·ha–1 nitrogen (N) using Super Rainbow 16N-1.8P6.6K plus minor elements (Agrium U.S. Inc., Denver, CO). An additional 28.5 kg·ha–1 N was applied in late August each year after planting using the product mentioned previously. Southeast-facing branches were flagged during winter 2007–08 and material used for
the duration of the study was collected from the same branches. Leaves were collected 8 Feb. 2008, 17 Aug. 2008, 9 Feb. 2009, and 5 May 2009 and frozen at –80 C until analysis. Temperature and precipitation data at the Tifton Campus for the duration of the study are included in Table 1. Supplemental irrigation was used only at the time of new plant establishment within plots. Chlorophyll and carotenoid extraction, analysis, and calculations. Three subsamples of leaf tissue were collected from the 12 individuals and Ca+b and Cx+c were extracted by grinding 85 mg leaf tissue three times in 3.33 mL 80% aqueous acetone and the extract was transferred to a test tube and brought to a final volume of 10 mL. After the third grind in acetone, the leaf material remaining was transferred to the test tube containing the extract and maintained in the dark at 4 C for 1 h to ensure complete extraction. Two milliliters of the extract was centrifuged for 30 s at 6800 gn. The supernatant was then transferred to a cuvette and absorbance was measured at 470 nm, 646 nm, and 663 nm using a GENESYSä 10 Spectrophotometer (Thermo Electron Corp., Madison, WI). Absorbance for all samples at each wavelength was between 0.2 and 0.8. Determination of Ca, Cb, and Cx+c was performed using calculations from Lichtenthaler and Wellburn (1983). Ca content was calculated using the formula: Ca (mg·L–1) = (12.25 · A663) – (2.79 · A646). Cb content was calculated using the formula: Cb (mg·L–1) = (22.5 · A646) – (5.1 · A663). Ca+b content was determined by summing Ca and Cb values. Total carotenoid content was determined using the formula: Cx+c (mg·L–1) = [1000 · A470 – (1.82 · Ca) – (85.02 · Cb)]/198. Chlorophyll and carotenoid contents were expressed in mg·g–1 of dry weight after being corrected for moisture content (MC) as follows: three unanalyzed leaf subsamples were
Table 1. Temperature and precipitation data at the University of Georgia Tifton Campusz for the duration of a study to evaluate pigments and color of Cryptomeria japonica from Nov. 2007 through Aug. 2009. Avg daily maximum Avg daily minimum Avg daily Month temp (C) temp (C) temp (C) Nov. 2007 20.8 7.4 14.1 Dec. 2007 19.4 8.0 13.7 Jan. 2008 14.0 3.9 8.9 Feb. 2008 18.9 5.6 12.2 Mar. 2008 20.9 8.5 14.7 Apr. 2008 24.1 13.0 18.6 May 2008 28.8 17.1 22.9 June 2008 32.5 21.0 26.7 July 2008 32.3 21.8 27.0 Aug. 2008 30.7 21.9 26.3 Sept. 2008 29.8 19.3 24.5 Oct. 2008 23.7 12.5 18.1 Nov. 2008 18.5 6.7 12.6 Dec. 2008 18.8 7.5 13.1 Jan. 2009 15.7 4.7 10.2 Feb. 2009 17.1 4.3 10.7 Mar. 2009 20.9 9.6 15.3 Apr. 2009 24.0 12.6 18.3 May 2009 27.6 18.6 23.1 June 2009 32.8 22.3 27.6 July 2009 31.7 21.6 26.6 Aug. 2009 31.5 22.0 26.7 z U.S. Department of Agriculture Zone 8b, lat. 3149# N, long. 8353# W.
HORTSCIENCE VOL. 48(12) DECEMBER 2013
Rainfall (cm) 2.3 14.4 8.7 19.9 8.1 7.7 4.7 5.8 8.4 31.3 0.4 16.3 9.9 7.9 4.2 5.5 20.9 21.0 16.6 5.0 17.6 17.8
collected to determine mean MC for each individual (replicate) for each harvest date using the formula {MC = [(initial weight – dry weight)/initial weight] · 100}. Color rating. Plants were observed within 1 week of the four leaf collection dates. Five evaluators assigned color ratings from 1 (very brown/yellow; off color) to 5 (very green). Ratings of 3.5 to 4 would be considered acceptable for landscape use. All plants were addressed from the southeast side, directly in front of flagged branches. Mean rating for each individual was calculated and used for statistical analysis. Design and statistical analysis. The experimental design used repeated measurements across winter and summer on each individual with subsamples taken on the individual branches. Relevant covariate information, chlorophyll and carotenoid measurements, was obtained for each subsample, whereas the target response, color rating, was characterized at the individual level. To assess the relationship between chlorophyll and carotenoid measurements with the observed color rating of the trees, data were analyzed using a mixed model framework (PROC MIXED, SAS Version 9.3; SAS Institute Inc., Cary, NC). Random intercept models (linear models with random components with each variant of tree having a different intercept, but the slopes are assumed the same) were constructed to assess the relationship between the chlorophyll and carotenoid measurements with the observed greenness of the trees. These models were of the form: Greenijk ¼ m þ treei þ seasonj þ b covariateijk þ Eijk , treei ; N ð0, st Þ, Eijk ; N ð0, sE Þ:
(1)
To determine an overall best model, Akaike’s Information Criterion (AIC) was used, in which a lower number represents a better fit. Also of interest in this research was to test for differences across the seasons in color ratings and chlorophyll and carotenoid easurements. For each measured pigment, a mixed-effects analysis of variance was constructed Pigmentijk ¼ m þ treei þ seasonj þ Eijk , treei ; N ð0, st Þ, Eijk ; N ð0, sE Þ
(2)
Hence, examining season term for each of these five models established whether there were seasonal differences in each pigment. The previous two statistical analyses aggregated the subsamples such that a single observation per tree per period was used. To assess whether summer greenness measures are predictive of winter greenness, a simple regression model was used, for which summer and winter greenness ratings were aggregated for each individual. Given the experimental constraints in which only a single tree from each taxon could be planted, there is no replication of taxa. Hence, statistical methods cannot be used to determine
1453
whether certain cultivars or varieties exhibit higher greenness ratings. Rather, our objective here was to use these named taxa to determine if Ca, Cb, Ca+b, or Cx+c can be used in a model to predict winter greenness for C. japonica. Results Ca+b (P = 0.0051), Cx+c (P = 0.0266), and the ratio of Ca+b to Cx+c (P = 0.0188) exhibited a positive relationship with greenness after accounting for season and tree effects (Table 2). In contrast, the ratio of Ca to Cb was not related to greenness. Using AIC, both Cx+c (AIC = 91.7) and Ca+b (AIC = 91.9) have superior model fits than ratio of Cx+c to Ca+b (AIC = 93.2). Similarly, visual color rating correlated with Ca+b (r = 0.535), total Cx+c (r = 0.375), and their ratio (r = 0.520) when pooled across years (Table 3). In addition, there was linear relationship between Ca+b and Cx+c regardless of season, summer (R2 = 0.94) and winter (R2 = 0.88), when pooled across 2 years (Fig. 1). The slope of the relationship (y = Cx+c; x = Ca+b) remained relatively consistent (4.47 and 5.05) across seasons; however, the intercept, a measure of Cx+c levels, decreased for 0.52 to –1.35 from summer to winter, respectively (Fig. 1) Each of the five attributes, observed greenness and measured chlorophyll and carotenoids, exhibited differences between summer and winter with higher values in the summer (Table 4). Using observed summer greenness pooled across years, winter greenness can be modeled (P = 0.017, R2 = 0.45) (Fig. 2). Discussion Mean Ca+b content was similar to previous values for japanese-cedar (Ida, 1981).
There were differences in mean Ca+b between seasons with a higher content in summer than winter when taxa were pooled (Table 4). Previous studies also have observed reduction in Ca+b content during winter in japanese-cedar (Han et al., 2003, 2004) and sitka spruce [Picea sitchensis (Bong.) Carr.] (Lewandowska and Jarvis, 1977). However, no increase in Ca+b or alteration of Ca:Cb ratio (discussed subsequently) was observed during spring recovery of Scots pine (Pinus sylvestris L.) from winter stress (Ottander ¨ quist, 1991). and O In Winter 2007–08, chlorophyll fluorescence was measured as an estimate of photosystem II (PSII) efficiency to assess damage. However, the values collected were extremely inconsistent (data not shown) and measurements were discontinued. Previous studies have successfully measured chlorophyll fluorescence (Han et al., 2003, 2004; Han and Mukai, 1999); however, the data we collected were not useful in evaluation of efficiency or damage to PSII. Therefore, we observed the relationship between Ca:Cb in an attempt to estimate the status of the photosynthetic apparatus. The ratio of Ca:Cb was lower in the winter than the summer (Table 4), which agrees with Han and Mukai (1999) who observed a slight decrease in the Ca:Cb ratio of japanese-cedar during the period in which chlorophyll was being lost (early winter). This also agrees with the findings of Wolf (1956) who observed this phenomenon in 25 tree species, although that study was conducted solely on angiosperms. The reverse was observed in sitka spruce, which showed a slight increase in Ca:Cb ratio during winter (Lewandowska and Jarvis, 1977). A decreased ratio of Ca:Cb indicates that PSII reaction centers are affected more than light-harvesting complexes, because Cb
Table 2. Mean total chlorophyll [C(a+b)], ratio of chlorophyll a (Ca):chlorophyll b (Cb), total carotenoid [C(x+c)], and visual color rating (Rating) of 12 replicates comprised of the same number of different taxa of Cryptomeria japonica evaluated at the University of Georgia Tifton Campusz during winters of 2007–08 and 2008–09 and summers of 2008 and 2009. Ca:Cbx Cx+cw Ca+b:Cx+c Season Ca+by Winter 2007–08 2.82 2.57 0.85 3.28 Summer 2008 3.98 2.78 0.83 4.83 Winter 2008–09 2.30 2.59 0.7 3.24 Summer 2009 4.8 2.81 0.91 5.28 z U.S. Department of Agriculture Zone 8b, lat. 3149# N, long. 8353# W. y Mean Ca+b in mg·g–1 dry weight. x Mean ratio of Ca:Cb. w Mean Cx+c in mg·g–1 dry weight. v Visual color rating based on color with 1 = yellow or brown and 5 = green.
Ratingv 2.98 3.23 2.56 3.60
Table 3. Correlation between total chlorophyll (Ca+b), ratio of chlorophyll a:chlorophyll b (Ca:Cb), total carotenoids (Cx+c), ratio of Ca+b:Cx+c, and visual color rating (Rating) of 12 replicates comprised of the same number of different taxa of Cryptomeria japonica evaluated at the University of Georgia Tifton Campusz during winters of 2007–08 and 2008–09 and summers of 2008 and 2009. Ca+b Ca:Cb Cx+c Ca+b:Cx+c 1.00 0.048 0.805*** 0.840*** Ca+b 1.00 –0.259 0.304* Ca:Cb 1.00 0.382** Cx+c 1.00 Ca+b:Cx+c Rating z U.S. Department of Agriculture Zone 8b, lat. 3149# N, long. 8353# W. *, **, *** indicate significance at P # 0.05, P # 0.01, and P < 0.0001, respectively.
1454
Rating 0.535*** 0.165 0.375** 0.520** 1.00
is found only in the light-harvesting complexes and Ca is part of the PSII reaction center (Hoober, 1998). Furthermore, characterization of the subunits of PSII reaction center has identified the D1 subunit’s involvement in photoinhibition (He and Malkin, 1998) and Sandmann and Scheer (1998) reported that pheophytin a (demetalated Ca) may be used as a marker for PSII. Differences were observed between seasons for Cx+c (Table 4). Robakowski (2005) observed a reduction in total carotenoid content in norway spruce [Picea abies (L.) Karst.] and mugo pine (Pinus mugo Turra) during winter but carotenoid content remained stable for silver fir (Abies alba Mill.). Differences were attributed to the latter species’ ability to maintain high photosynthetic capacity during winter as its mechanism for preventing oxidative stress (Robakowski, 2005). We measured total carotenoid content in japanese-cedar leaves, whereas previous studies separated carotenoids (Han et al., 2003, 2004; Han and Mukai, 1999). However, the value for yellow carotenoids presented by Ida (1981) appears comparable to our values. A positive, linear relationship was found between Ca+b and Cx+c in the current study in all seasons (Table 3). When data were pooled over 36 taxa belonging to Gymnospermae, Ida (1981) found a similar relationship. This relationship seems counterintuitive because chlorophyll decreases and carotenoids increase during winter (Han et al., 2003, 2004). Oserkowsky (1932) reported data that appeared to fit a positive linear relationship between yellow pigments (xanthophyll) and chlorophyll of hardy pear (Pyrus communis L.). That study cites other research (Sj¨oberg, 1931; von Euler and Hellstr¨om, 1929) that reported a simultaneous increase in green and yellow pigments in etiolated seedlings. The physiological basis for the concomitant increase in chlorophyll and carotenoids has been established as a means to prevent photooxidative damage resulting from ROS (Lambers et al., 1998). Excess energy is present, particularly during winter, because light capture and energy transfer occur more rapidly than the downstream biochemical reactions. However, carotenoids accumulate during winter and prevent damage by dissipating this excess energy as heat (Lambers et al., 1998). There was a statistically significant relationship between both Ca+b and Cx+c content with visual rating. The hypothesis at the beginning of the experiment was that Ca+b content may be used as a predictor of eventual field performance; the higher Ca+b content, the more green the plants would remain in winter. We failed to reject this hypothesis. Previous studies have observed linear (Marquard and Tipton, 1987; Yadava, 1986) and quadratic (Netto et al., 2005) relationships between SPAD readings [SPAD-501 (Marquard and Tipton, 1987; Yadava, 1986); SPAD-502 (Netto et al., 2005)] and total chlorophyll content determined spectrophotometrically. In our study, Cx+c also had a predictive relationship with observed greenness. This may be explained by the simultaneous increase in HORTSCIENCE VOL. 48(12) DECEMBER 2013
Fig. 1. Linear regression of total chlorophyll content over total carotenoid content in summer (S) or winter (W), pooled across 2 years, 12 replicates comprised of the same number of different taxa of Cryptomeria japonica evaluated at the University of Georgia Tifton Campus (U.S. Department of Agriculture Zone 8b, lat. 3149# N, long. 8353# W). The values for S and W are means of the 2 years.
Table 4. A mixed effects analysis of variancez (ANOVA) table to evaluate seasonal differences, summer minus winter, of total chlorophyll (Ca+b), ratio of chlorophyll a:chlorophyll b (Ca:Cb), total carotenoids (Cx+c), ratio of Ca+b:Cx+c, and visual color rating (Rating) of 12 replicates comprised of the same number of different taxa of Cryptomeria japonica evaluated at the University of Georgia Tifton Campusy pooled across years 2007– 09 (winters) and 2008–09 (summers). SE Covariate Estimate 1.8260 0.2078 Ca+bx 0.2144 0.04233 Ca:Cb 0.09274 0.03953 Cx+cw 1.7999 0.1158 C(a+b):C(x+c) Rating 0.6167 0.1449 z Pigmentijk ¼ m þ treei þ seasonj þ Eijk , treei ; N ð0, st Þ, Eijk ; N ð0, sE Þ y U.S. Department of Agriculture Zone 8b, lat. 3149# N, long. 8353# W. x Mean Ca+b in mg·g–1 dry weight. w Mean Cx+c in mg·g–1 dry weight.
P
